WE ARE UNDER CONSTRUCTION - DON'T MIND THE DUST!

News media must present more reliable research

By Emilio Salvia

After attending a fraternity conference this past weekend and taking a statistics course this term, I’ve thought a lot about the consequences of poorly conducted research studies.

From the news to the classroom, even to Facebook, research studies that influence the way we think, believe and feel are everywhere. There are innumerable amounts of laughable and mediocre scientific studies out there, the kind of stuff that Fox News would love and would probably use to appease their agendas and buddies.

A good example of an invalid scientific study earning undue credibility was the study that established a link between nonviolent TV programs and increased aggression in a group of boys. One group could watch what they wanted while the other could not watch any violent TV.

As it turns out, the surprising link between non-violent programs and aggression was just illusory. The boys who had their TV regulated and didn’t get to watch what they wanted showed a significant increase in aggressive personality. This was simply because they were upset that their friends could watch what they wanted and they couldn’t.

Another example—and a personal favorite—is the study that linked drinking coffee to increased rates of lung cancer in individuals. At the end of the study, the researchers saw a reasonable correlation between those who drank coffee and those with cancer. They concluded that the two variables are linked.

“Oh, hey!” likely said the researchers after the study was published. “Maybe the correlation is due to second hand smoke in coffee shops!” Classic.

Why talk about this though? These studies are obviously extreme examples, you say. That’s true, but, as it turns out, other mediocre scientific studies are plagued by the problems of confounding variables, poor sampling, experimenter bias and other, more nuanced problems like accidental scientific misconduct.

What’s the result? Wasting considerable amounts of funding for near-useless studies, lessening the public’s support for science, fueling stereotypes and causing unneeded unrest and worry.

The aspects of stereotyping and wasted funds are pretty important, at least to me. Many scientific studies are invalid because of the existence of stereotypes—expressed through result expectations and biases. These studies tarnish science’s credibility and, in doing so, waste limited funds. Similarly, poorly conducted studies can lead to lovely journalism articles with headlines like “Higher Incidence of ‘Variable x’ Among Fraternity Men.”

I am not trying to discredit studies or be naïve about the ones I don’t like. It’s just that sometimes the studies we are presented with don’t have much validity. But, at other times, responsibly conducted studies can offer a reality check. I cannot go into great detail about any of the things I’ve talked about so far in this article, but you get the idea.